Spaceborne GPSCurrent Status and Future Visions

Frank H. Bauer, Kate Hartman and E. Glenn Lightsey

NASA Goddard Space Flight Center

Greenbelt, Maryland, 20771301-286-8496

frank, bauer@gsfc, nasa. gov

ABSTRACT

The Global Positioning System (GPS), developed by the

Department of Defense, is quickly revolutionizing thearchitecture of future spacecraft and spacecraft systems.

Significant savings in spacecraft life cycle cost, in

power, and in mass can be realized by exploiting Global

Positioning System (GPS) technology in spacebomevehicles. These savings are realized because GPS is a

systems sensor--it combines the ability to sense spacevehicle trajectory, attitude, time, and relative ranging

between vehicles into one package. As a result, a

reduced spacecraft sensor complement can be employed

on spacecraft and significant reductions in space vehicle

operations cost can be realized through enhanced on-

board autonomy. This paper provides an overview of the

current status of spaceborne GPS, a description of

spaceborne GPS receivers available now and in the near

future, a description of the 1997-1999 GPS flight

experiments and the spaceborne GPS team's vision forthe future.

INTRODUCTION

GPS technology holds great promise for terrestrial aswell as space-based users. The world is just beginning to

understand the tremendous benefits and great potential

that this technology can deliver to the military and

civilian transportation industry. Safer air travel,

improvements in search and rescue systems, improved

Earthquake monitoring, tractor-trailer tracking and

enhanced farming techniques are just some of the

terrestrial-based spinoffs from this technology.Handheld GPS receivers are now available for less than

$100 and millions are being sold each year. The benefits

of this technology on Earth is extensive. The benefits

are equally extensive for spacecraft and space systems.

Significant reductions in spacecraft costs, improvements

in spacecraft autonomy and new, exciting scientific

opportunities can be accomplished through the infusionof this technology on spacecraft and spacecraftconstellations of the future.

While this technology holds great promise, its

incorporation on spacecraft has been delayed for several

reasons. See figure 1. The tremendous success of the

very lucrative terrestrial GPS market has, in fact, stifled

the development of spaceborne GPS receivers.

Companies with GPS expertise are more interested in thelucrative terrestrial-based GPS market. They are not

interested in diverting their GPS talent on the relatively

small space-based market. This has restricted the

development of spaceborne receivers to meet thedemands of future spacecraft requirements.

In addition to the above problem, there are many

technical challenges that must be overcome before

space-borne GPS bears all its fruit. GPS receivers thatare used in space are very different than their terrestrial-

based cousins. The high speeds of low Earth orbiting

spacecraft result in signal doppler and doppler rate which

are significantly higher than what is observed on the

ground. Also, the GPS satellites rise and set on low Earth

orbiting vehicles much faster than terrestrial based users

(approximately 45-50 minutes versus 6 hours). Thesedifferences result in a significantly larger search space

for spaceborne GPS requiring a much faster solution.

Figure 1 outlines these problems and describes the

technological "stairsteps" that the spaceborne GPScommunity needs to climb before spacebome GPS has

fully matured.

The spaceborne GPS community must expediently

overcome the GPS technology hurdles shown in figure I

and develop a stable of robust spaceborne GPS receivers

that meet future mission requirements. To this end, the

GPS team at the Goddard Space Flight Center (GSFC)

has been fostering government/university/industry

partnerships in spaceborne GPS technology. The

objectives of these partnerships are to promote the

development and use of spaceborne GPS through a four-

pronged program. The elements of this program include:

I) The development of spaceborne GPS receiverswhich satisfy the future spacecraft mission

requirements.

2) The development of the techniques required to

integrate this technology on spacecraft in a costeffective manner.

3) The validation of this technology through a series offlight experiments; and,

4) The development of enhanced autonomy techniquessuch as autonomous orbit control and formation

flying.

The GPS technology program being accomplished at

NASA Goddard encompasses the use of GPS primarily

as an engineering sensor as compared to using GPS for

science measurements. As an engineering sensor, GPSdetermines spacecraft attitude, relative and absolute orbit

position, and time. When used as a science instrument,

GPS performs gravity, atmospheric sounding, oceanreflection, and ionospheric sounding measurements.

While the science and engineering aspects of using GPS

in space follow somewhat similar hardware developmentpaths, the receiver robustness, data requirements, and

receiver operation in an engineering application is verydifferent from its use as a science instrument. This is

primarily because the engineering function must

maintain space vehicle health and is geared towards

reducing space mission costs through reduced sensor

complements and enhanced vehicle autonomy. To

achieve these objectives, the GPS receiver must provide

real-time, autonomous, onboard support to the

spacecraft. This contrasts with the objectives of GPS

science applications which demand extensive postprocessing of the data to glean as much information from

the science data as possible. This data reduction is

currently performed by a cadre of ground operationspersonnel. Moreover, the GPS science instrument is not

required to maintain mission critical functions; thus, theelectronics hardware embedded in the receiver and the

Roadblocks Impeding

Space-Based GPS Technology

• Percel_on that Gilts technology is mature and requi¢_ no res_wch

am/devdopment effort to fly in SlatCe

• More lucrative GPS receiver mart_ for terrestrial applications

• Differences _n space-lmsed & gromai-lmed GPS reception

• Require numermm flight expeg4.nwnts to _climb the technological

stairsteps"poinltq - Stellar p_inhlqg - _4niq

- AnttadeDele_miuaam -AnimdeCaanml -Malapath_

- AIIIII_mOIS OrlM| l]_f. - P,_lLlllllve R_rill_ - AIIl_l_llOIIII Oriel _111_'_

- Farmllos Ityialg - GEOdtEO Opa - I_'_M,e Tlmmll

- Scmo¢ ralmamms - TYdRX capa_|ty - Uuldtd eald start

Figure 1: Roadblocks to Space-Based GPS

algorithms used to obtain the GPS data can be verydifferent from a GPS engineering device.

VISION FOR SPACEBORNE GPS

The current vision for spaceborne GPS technology is

encapsulated in figure 2.

Space-Based GPS Technology:

Vision of Highly Autonomous Spacecraft

• Vision: Improve space vehicle

autonomy; Reduce design and

operations costs through theinfusion of new GPS hardware and

sensing techniques

• GPS Sensing Capabilities:

• Ataonomous Orbit Determination

4. Accurate Time Synchronization

+ Coarse Attitude Det_ion

+ Accurate Relative Ranging

Between Platforms

• 21st Century Engineering

Applications:

4. Autonomous Onboard Navagatiort.

Operations, and Orbit Control

• Attitude Delerrah'talion and Control

• Rendezvous & Proximity Operations

÷ Formation Flying/Coorditlal_l

Platforms

÷ Micrmatellit_ w/3 _is control

• GPS Perfornrmnce at GEO and Above

• Governrc_t. University. and

Commercial Partnership

Figure 2: Spaeeborne GPS Vision

As shown above, the vision embodies the exploitation of

GPS as a systems sensor----capable of determining

spacecraft trajectories autonomously, delivering precise

time synchronization to spacecraft electronics, sensing

vehicle attitude and measuring the relative distancebetween space vehicles. Achieving the vision outlined in

figure 2 will take many years to complete. The current

generation of receivers, being installed on today's flight

experiments, will definitely open new doors in space

vehicle design and autonomy. However, they will not

fully achieve the revolutionary changes in spacecraft

sensing and autonomy outlined in the vision statement

depicted in figure 2. The GPS technology roadmap,

shown in figure 3, illustrates that this will require

approximately 10 years of concentrated development

effort and probably three generations of flight receiver

designs--the first generation of which is available today.

The current generation (first generation) of engineeringreceivers have been installed on several spacecraft and

are being put through a battery of tests to validate their

on-orbit performance and to understand their limitations.

The information gleaned from these tests will be used to

give GPS users the information they need to successfully

integrate these current generation receivers onto their

spacecraft and will also give the GPS development team

the knowledge required to improve future receiverdesigns.

The second generation of receivers, expected to be

available at the beginning of the next millennium, will

be miniaturized versions of their predecessors, fabricated

1996 2001 2006

• Radiation Hardened

•Algorithms adapted for space

•Navigation (500 m Real Time,60 m Post Process)•Precise Time through Bus (2ms)

Commercial]

L I

•Robust algorithms•Modular S/W•Miniaturized H/W(GPS on Board)• Enhanced initialization

• Navigation (5m WAAS)• Relative Navigation (10m)•Ops above the constellation

•TX/RX Capability•GPS on a Chip•Cold start w/o Ephemeris•Dual Freq. Miniaturized

•Attitude Determination (0.25-0.1 °) .Autonomous Orbit Control

LGlobalstarJ[Spa.anLiteI LCOMSATJ ISMEX/MIDEXl [

Figure 3: GPS Receiver Technology Roadmap

Key

on multi-chip modules. They will include modular

object oriented software and robust algorithms to

significantly improve their operation in space. The

receiver will be capable of quick (5 minute) cold start

initialization without ground intervention. This new

receiver set will enable real-time space vehicle

navigation to meter performance levels using the WideArea Augmentation System (WAAS) and will provide

relative navigation knowledge to 10 meters.

The third generation of spaceborne receivers will enable

inexpensive micro-spacecraft. Electronics technology

improvements will allow the third generation receiver to

be further miniaturized. A transmit/receive capability

will be included in this generation receiver to enable

autonomous formation flying. The "GPS-on-a-Chip"

vision, proposed in 1995 by Goddard, Stanford

University, and JPL, should be fully realized with this

receiver. Robust use of GPS in Geostationary Earth Orbit

(GEO) and High Earth Orbit (HEO) will be realized withthis generation of receivers.

•GEO & HEO Applications•Relative Navigation (5 cm)•Formation Flying•Microsat Constellations

TDRSS ]

GOES ]

GPS-AN ENABLING TECHNOLOGY

Incorporation of GPS on space vehicles enables three

primary technologies: space vehicle autonomy, ground

station automation and virtual spacecraft constellations.

Without GPS, all three of these technologies could grow

and mature over time--but not without a high ground

operations cost penalty and long-term developmentschedules. With GPS, each of these technologies become

spacecraft breakthrough technologies, providing low

cost, attractive opportunities for missions of the future.

Space Vehicle Autonomy--Low cost autonomous

navigation, on-board maneuver planning andautonomous constellation control all become feasible

when GPS is employed. Traditionally, spacecraft

navigation would be accomplished on the ground

through ranging and trajectory determination techniques.

Planning and controlling the orbit of a single spacecraft

from the ground is labor intensive. Performing these

functions from the ground on several spacecraft

simultaneously is extremely complex and introduces an

overwhelming ground personnel requirement. This

would place a considerable burden on the ground

operationsteamwho mustnot only ensurethatallspacecraftmaintainproperorbitalspacing,but alsomonitorspacecraftstatesof healthwhilecollectingallpertinentdata requiredto achieveoverallmissionsuccess.The time, orbit and attitude data, obtained from

GPS, enables spacecraft system developers to

accomplish autonomous orbit maneuver planning and

autonomous stationkeeping maneuvers on-board the

spacecraft. This results in a substantial reduction in

mission operations costs.

Other spacecraft autonomy technologies enabled using

GPS are low cost, standardized spacecraft timing

systems through the spacecraft data bus, vehicle attitudedetermination and attitude control, and autonomous data

transmissions over ground stations. Miniaturized copiesof a GPS receiver can also serve as the heart of an

autonomous micro-sciencecraft providing attitude and

orbit sensing, attitude commanding, orbit control,

command and data handling services, and science

instrument timing all in one package.

Virtual Platforms--- Developing the technology to

produce virtual platforms will be a long range challenge.Similar to its enabler--GPS technology--there are

several technological hurdles that must be overcome to

go from autonomous navigation and constellation control

to one and two way lbrmation flying and finally to

virtual platforms (see figure 4). Despite these hurdles,

the Earth and space science community has already

started defining the new science measurements and

science missions that can be performed using clusters of

spacecraft flying in formation in virtual platforms.

One technological hurdle critical to enable autonomous

formation flying is the (transmit/receive) space vehicle

crosslink system that is to be incorporated in the third

generation GPS receiver (see figure 3). When the third

generation receiver is coupled with autonomous on-

board maneuver planning and orbit control software,

autonomous relative ranging between space vehicles and

formation flying becomes feasible. Using the crosslink-

GPS receiver/transmitter as an enabling step, relative

ranging and formation flying technology is expected to

revolutionize both manned and unmanned space vehicle

operations in the near future.

Once formation flying techniques are perfected, the

space and Earth science communities can perform new

and exciting missions unimaginable just a few years ago.

In the future, autonomous Space Shuttle and Space

Station rendezvous and docking using GPS will becomecommonplace. Very low cost scientific payloads, such

as Spartan, will be deployed from Space Station, fly in

Closed Loop Navigation

Autonomous Constellation

ControlAutonomous Formation Flying

of Multiple Satellites

1996 1999 2001 2006 2010

_IGINS- SEC-

On Board Autonomous 1- Way 2- Way True VirtualManeuver Constellation Formation Formation Formation PlatformPlanning Control F_ng Fbo_ f b_

Real Time Real Time Real Time Chase S/C Chase& Multiple S/C Na_gation &On Board On Board On Board reacts to Tanget SIC Na_gatir_ Attitude

Orbit _ V AV data f_om react in collective/X & reactingDetermination Determination Execution Target S/C concert autonomously collectively

Figure 4: Autonomous Onboard Orbit Determination and Control

formation, and autonomously return for eventual

retrieval. In the process, these enabling technologies

will allow synchronous science measurements to occur

on multiple space vehicles. Multiple spacecraft

formation flying as a virtual platform and gatheringconcurrent science will soon be feasible once the GPS

and tbrmation flying technologies have fully matured.

Ground station automation--The ultimate goal in

ground station automation is to develop true "LightsOut" operations. In other words, operating spacecraft

with a small team that is only present on the first

(daylight) shift. Since the flight and ground segments are

intertwined, spaceborne GPS and GPS on the ground are

required to enable lower cost "Lights Out" systems. As a

result, GPS receivers are being incorporated in ground

stations as a timing synchronization source. Ground

operations personnel and specific ground passes can be

reduced since ranging and Ephemeris uplinks are no

longer required. Ground-based formation flying would

require a substantial ground personnel requirement. This

is not needed if autonomous formation flying is

employed.

18" DSS DishGPS Attitude

"GPS Attitude Array D,.' Receiver

AIt / Az RotatorRotator Controller

Antenna System Rotatori Interface

DSS Receiver: and Monitor Computer

Client Component Support Equipment

Further automation of ground stations can occur through

the use of systems such as the Transportable Antenna

Pointing System (TAPS). In 1994, engineers at GSFC

proposed the development of a Transportable Antenna

Pointing System (TAPS) using GPS. The TAPS concept

integrates a GPS attitude and navigation receiver into a

spacecraft antenna pointing system to provide complete

autonomy in a ground station operation (see figures 5

and 6). Using GPS, the ground station receives precise

time synchronization, 3 dimensional real-time positioninformation (Latitude, Longitude and Height) and

attitude sensing relative to an Earth-Centered-Earth-

Fixed coordinate frame--all required to accomplish

autonomous antenna pointing to spacecraft. TAPS can

also be mounted on a moving vehicle and provide

precise antenna pointing despite changes in vehicle

direction or vehicle attitude as long as a clear field of

view to the satellite is provided. Thus, mobile or portable

satellite operations can be accomplished on ships, vans,

balloons or airplanes or point small instruments on Space

Shuttle Hitchhiker missions. This concept was

successfully demonstrated in 1996.

CURRENT AND FUTURE SPACEBORNE GPS

RECEIVERS

The first generation of GPS receivers available to the

space community are summarized in Table 1. These

receivers are grouped according to space heritage.

Front View Rear View

Figure 5: TAPS

Figure 6: TAPS Block Diagram

Following the figure is a top-level description of the

spaceborne receiver's heritage, radiation hardness, and

specific capabilities. Spaceborne GPS users should be

cautioned that these are all first generation GPS

receivers. As such, they are not totally robust. Many areknown to possess some algorithmic issues, software

bugs, hardware constraints and data outages. These

issues are predominantly not GPS manufacturing or

design problems. They are more a reflection of the

maturity of the technology. They are expected to becorrected in future receiver software releases or when the

next set of GPS receivers are available.

Name

Receivers Flown in Space

Table I: Spaceborne GPS Receivers

Channels Freq # Antennas Capabilities

Motorola Viceroy 12 Single

Motorola Monarch 12 Dual P(Y)

Collins MAGR/S 5 Dual P(Y)

SS/L Tensor 9 Single

AOA Turbostar 8 Dual P

Trimble TANS Vector 6 Single

Current Developments

Honeywell/Trimble/GSFC SIGI 12 Single

GSFC PiVoT 24 Single

SSTL GPS 12/24 Single

JPL/GSFC/Stanford University 12 Dual

GPS on a Chip

2

2

I

4

!

4

Navigation, Timing

Filtered Navigation, Timing, Rad Hardened

Navigation, Timing

Attitude, Filtered Nav, Timing, Rad Hardened

Precision Nav, Timing, P & P Codeless

Attitude, Navigation, Timing

4

2/4

4/5

4

Attitude, Navigation, Timing

Filtered Nav, Timing, (Attitude Option)

Attitude, Filtered Navigation, Timing

Precision Navigation, Timing, Attitude

Currently Available Spaceborne Receivers

Motorola Viceroy

The Viceroy can now be considered a heritage receiver.It has flown on MSTI-3, Seastar and on a Mir

experiment. It is also expected to fly on several future

missions, including Earthwatch. As shown in the figure,

it is a basic C/A code navigation & timing receiver. Its

electronics were not designed for radiation hardness, but

it can withstand a moderate (15 Krad) environment

which is adequate for most low Earth Orbiting missions.

Motorola Monarch

The Monarch is Motorola's premier GPS receiver. It is a

12 channel, radiation hardened (> 100 Krad) receiver that

is capable of receiving the encrypted P(Y) code.

Developed primarily to meet DoD requirements, it can

output data in RS-422, MIL-STD 1553 and Remote

Interface Unit (RIU) formats. Heritage for this receiver

is derived from the P code dual frequency receivers

flown on TOPEX and the Extreme Ultraviolet Explorer

spacecraft. The receiver has an embedded 1750A

microprocessor and the receiver software is written inAda.

Rockwell/Collins MAGR/S

The Miniature Airborne GPS Receiver/Shuttle

(MAGR/S) is currently the Space Shuttle receiver [1].

This 5 channel receiver is used to provide the Shuttle

with real-time navigation and timing. It will eventually

replace the use of TACAN during the landing phase of

the mission. This is a P(Y) code receiver and NASA

Johnson Space Center is using the receiver in its fully

secure mode. The receiver is expected to be placed in

primary service (the primary Shuttle navigation sensor)in 1999.

SS/L Tensor

The Space Systems/Loral Tensor [2][3] is a radiation

hardened (100 kRad) attitude, orbit and timing receiver.

It can accommodate up to four antennas and multiplexesthe 9 channel receiver through all four channels to

determine vehicle attitude at sample rates up to 10 Hz.

A navigation filter is included as a commanded option to

improve navigation performance to approximately 100meters. This receiver has flown on the SSTI-Lewis

spacecraft and is expected to fly on the TRW ROCSAT

spacecraft, the GSFC EO-I spacecraft and the SS/LGlobalstar constellation.

Allen Osborne Associates Turbostar

The Turbostar GPS receiver was developed jointly by

JPL and Allen Osborne Associates to support NASAscientific investigations in space. This 8 channel, dual

frequency P code receiver has a high (50 Hz) sample rate

that enables the device to perform atmospheric and

ionospheric sounding measurements. Using cross

correlation techniques, the Turbostar can support P-

codeless operation. Similar to the TANS Vector, theTurbostar is a commercial receiver that has been

ruggedized for space. The Turbostar has flown on the

GPS-METandWakeshield2& 3. It isexpectedto flyon theDanishORSTED,theSouthAfricanSUNSATandtheU.S.GeosatFollow-on(GFO)satellites.UsingtheInternationalGPSServicefor Geodynamics(IGS)globalnetworkof GPSreceivers,theTurbostar receiver

can achieve sub-meter post-process orbit determination

performance.

Trimble TANS Vector

The Trimble TANS Vector is a commercial terrestrial

receiver that has been ruggedized for use in space.

Spaceborne software was developed for this receiver

through a partnership with engineers from Trimble,

Stanford University, and GSFC. Using this software, the

receiver provides 3-axis attitude to approximately 0.5

degree [4][5], navigation to 450 meters and timing to 0.1

microsecond. Approximately a dozen receivers haveflown on missions such as Orbcomm, Crista SPAS,

REX-II, GADACS, GANE as well as several Space

Shuttle Experiments. The receiver is somewhat radiation

soft with a tolerance of approximately 8 kRad. Despite

this, the receiver has supported low Earth orbiting

missions for well over a year and in some cases 3 years.

Current Developments for Space

The current spaceborne GPS receivers available

commercially have not kept pace with NASA's current

and future requirements. NASA needs receivers that arelow cost ($100K or less), low power and can be

reconfigured to meet the specific mission requirements.None of the low cost receivers currently available meet

the demanding real-time navigation requirements (<100

meters) of many Earth science missions. All the

spacebome receivers have been derived from terrestrial

receivers, inheriting the zenith-pointing assumptions that

limit the robustness required in spaceborne applications.Therefore, NASA has initiated several in-house GPS

development initiatives to bring spaceborne GPS to

spacecraft users. These, as well as the low-cost GPS

receiver development at Surrey Satellite TechnologyLimited are described below.

GSFC PiVoT

An in-house NASA GSFC development is called PiVoT,which stands for Position, Velocity, Time. This modular

receiver design is based on the GEC Plessey Chipset

[6][7]. PiVoT is expected to be able to withstand a

moderate (15-20 kRad) radiation environment. Robust

algorithms embedded in PiVoT will allow the receiver to

operate in any vehicle orientation and support the

demanding navigation requirements of 95% of NASA's

Earth science and space science missions. The design

will support up to 4 antennas and an attitude sensing

option is planned. The protoflight PiVoT receiver will

be delivered in 5/98 with protoflight qualification

completed around 10/98.

SIG! Honeywell/Trim ble/Collins/GSFC

The Space Integrated GPS/INS

Inertial Navigation System for

NASA's Space Shuttle,

International Space Station andCrew Return Vehicle. The INS

consists of a Ring Laser Gyro,GPS receiver, inertial

navigation computer and power

supply mounted in a standard

Embedded GPS INS housing.

(SIGI) represents a new

Figure 7: SIGi

See figure 7. Two spacebome receivers are being

produced. The first is the Collins GEM dual frequency

receiver. It is both C/A and P(Y') code capable and

supports navigation and timing functions. This receiver

will be embedded in the SIGI that flies on the Space

Shuttle. The second receiver is being developed through

a partnership between Honeywell, Trimble and NASA

Goddard. This receiver will support the International

Space Station and Crew Return Vehicle. The Trimble

receiver used in this application supports attitude,

navigation and timing. The receiver can track up to 12

satellites. It is an upgrade of the successful TANS

Vector receiver. The attitude algorithms and softwareembedded in this receiver are being developed by NASAGSFC.

JPL/GSFC/Stanford GPS on a Chip

The GPS-on-a-Chip development effort is a joint,

collaborative program with JPL, Stanford University and

NASA GSFC as partners. JPL is developing the GPS-

on-a-Chip hardware and Stanford and Goddard will

develop the algorithms, software and validate the system

performance through ground-based testing. The

development approach was to develop the receiver

requirements, design and develop the prototype receiverand make the design available to future industrial

partners. The Phase I prototype receiver is expected to be

completed by the year 2000. The GPS on a Chip project

started in 1996. GPS-on-a-Chip is being designed to be

modular with an open architecture hardware and

software design and developed to be flight qualified in

the future. GPS-on-a-Chip will support multiple

antennas, can be used as a dual frequency P code

receiver or a single frequency C/A code receiver. It is

expected to support attitude, filtered navigation and

timing. Prototypes of this receiver are planned to be

flown on the SAC-C (1999), Champ (1999), and theGRAC E (2001 ) missions.

Surrey Satellite

Surrey Satellite Technology Limited (SSTL) is also

developing a GPS receiver based on the GEC Plessey

chipset [6][7]. Similar to the PiVoT design, the SSTL

receiver will incorporate radiation protection. The

receiver uses an ARM60 32 bit RISC microprocessor.

This receiver will support an on-orbit software uplink, a

feature that is lacking (but should be required) in many

commercial GPS receivers. Data is output via a CAN

interface, which is an automobile standard, similar to

RS-422. The receiver is being incrementally upgraded

and flown as an experiment on several Microsats. The

first flight is expected on TMSAT in late 1997 or Early

1998. TMSAT is a Technology Transfer Micro-satellite

being constructed for Thailand. This flight will include

the receiver in a 2 antenna configuration. The first flightof the 4/5 antenna receiver will be on UoSAT-12 which

is expected to fly in 1998.

1997-1999 FLIGHT EXPERIMENT OVERVIEW

The experiments accomplished prior to 1997 and those

planned for 1997-1999 expect to validate some of the

initial concepts of using GPS for spacecraft engineering

functions and to climb the technological stairsteps

depicted in figure 1. Each experiment is unique,

providing a wealth of information to the spaceborne GPS

receiver design community. These experiments will

provide valuable information on using GPS as an

autonomous navigation instrument, as a spacecraft

attitude determination sensor, as a precise time

synchronization a source, and as a relative rangingsensing device.

To date, two spacecraft GPS experiments have flown in

1997. Table 2 outlines the eight Goddard-collaborated

spaceborne GPS experiments scheduled for 1997-1999.

The table includes the mission name, the spacecraft, the

planned experiment objectives, the type of GPS receiver

(or receivers) to be flown and the launch date. After

reviewing these collaborative experiments, one may

observe that there is some overlap in some of themissions. It should be noted that this was done

intentionally. The GPS team's intent is to obtainexperimental data which is critical to factor into future

GPS receiver designs in a timely fashion despite launch

vehicle, spacecraft, or experiment failures that are part of

the risk of flying in space. The following paragraphs

provide more details on each experiment.

GPS Attitude Determination Flyer (GADFLY)

Experiment on the SSTI-Lewis Spacecraft

The Global Positioning System (GPS) Attitude

Determination Flyer (GADFLY) [8] experiment was

launched on the Small Satellite Technology Initiative

(SSTI) Lewis spacecraft in August 22, 1997 (see figure

8). The primary objective of the GADFLY experiment

was to demonstrate and validate the cost-saving, systems

engineering features that can be exploited by using GPS

receivers in space vehicles. The experiment's physical

hardware included four GPS antennas and pre-amplifiers,

cross-strapped to two Space Systems/Loral GPS Tensor

receivers. This was expected to be the first long-term,

on-orbit flight of a fully space-qualified GPS receiver

capable of simultaneously sensing space vehicle attitude,

orbit and providing a precise time reference.

The Space Systems/Loral Tensor receivers were flown

on GADFLY primarily as an attitude determination

experiment. However, an integral part of this experiment

was to demonstrate precise time distribution and provide

autonomous real-time navigation solutions to the

spacecraft subsystems and other experiments. The

spacecraft clock used a timing signal and a "time at the

tone" message from the GPS Tensor to maintain a timing

accuracy to better than two milliseconds. A GPS timing

testbed, developed at the Goddard Space Flight Center,was used to test the techniques required to transfer the

GPS timing information from the GPS Tensor onto aMIL-STD-1553 spacecraft bus. In addition to the timing

information, navigation solutions provided by the GPS

Tensor was intended to support the Lewis science

objectives. To improve the real-time navigation accuracy

delivered by the GPS Standard Positioning Service, the

GPS Enhanced Orbit Determination Experiment

(GEODE) [9] was being developed for GADFLY.GEODE, an enhanced navigation filter developed at the

Figure 8: GADFLY on SSTI-Lewis

Goddard Space Flight Center, is expected to provide

navigation accuracy on the order of l0 meters 16. The

GADFLY performance goals are shown in table 2.

The GADFLY team includes members from NASA

Goddard Space Flight Center, TRW (the SSTI-Lewis

spacecraft manufacturer), Space Systems/Loral, NASA

Johnson Space Center, Stanford University, Computer

Sciences Corporation, and Orbital.

Table 2: GADFLY Performance Goals

Attitude Determinatio®

Orbit Determination

[Spacecraft Reqniremenl

None (using GPS)

150 m 30 in-track

150 m 3a cross-track

230 m 30. radial

Time Tags: 2 reset.

I Hz update

GADFLY Goals

0.45 ° 30"

450 m 30" unfiltered

50 m 30" Tensor filtered

60 m 30" GEODE

Time Tags: < 1 mse¢

I Hz update

Precise Timing I msec, I Hz update Time Tags < I msec

Reference Discrete Pulse < I lasec

Table 3:1997-1999 Space-Based Experiments

Experiment Carrier Goals Receiver Date

GADFLY SSTI-Lewis Filtered OD, SS/L Tensor 11/5/97Time, RT AD

Seastar GPS Seastar

Earthwatch GPS Earthwatch/Earlybird

RT OD, Timing

RT AD, Timing,RT Precise OD

SHUCS STS-91/SPACEHAB RT AD, RT OD, Ant.

Pointing, Time

Motorola Viceroy Xx/xx

Trimble TANS Vector 12/14/97

Motorola Viceroy

Trimble TANS Vector 5/98

Clark-GPS SSTI-Clark RT AD, Filtered OD Trimble TANS Vector 6/98

AMSAT GPS AMSAT Phase IIID RT AD, Ops above Trimble TANS Vector 7/98GPS Constellation

SAC-A GPS SAC-A RT AD, RT OD, Trimble TANS Vector 7/98

Spinning AD

Enhanced Formation EO-I RT OD (20 m), Time, SSFL Tensor 5/99Flying AOC, AFF

GPS on the Shuttle All Shuttles RT OD Collins MAGR/S Numerous

KEY: RT=Real Time, OD=Orbit Determination, AD=Attitude Determination, AOC=Autonomous Orbit Control,AFF=Autonomous Formation Flying Techniques

SSTI-Lewis Mission Results

Prior to the mission, the GADFLY experiment was

subjected to a battery of tests to validate that the Tensor

receiver was ready for flight. These were successfully

completed and the GADFLY team was confident that the

experiment would provide and outstanding data return.During the first ground station pass, the Tensor receiver

was activated, acquired and locked on to sufficient

satellites to produce a navigation and time solutionwithin 20 minutes. The receiver remained active until

August 26 when the spacecraft was observed to be

spinning at a 2 rpm rate. This unrelated attitude control

problem resulted in the total loss of the Lewis mission

and a total loss in the GADFLY experiment. Despite

this unfortunate setback, the GPS community benefited

significantly from the GADFLY experiment. Some of

the experiment highlights include: 1) the flight

qualification of the Tensor--the first radiation hardened

receiver capable of determining attitude, orbit and time;2) the on-orbit demonstration of the Tensor reciever for a

brief period (several days) in a navigation and timing

configuration; 3) the development of the GEODEsoftware which is expected to serve over 95% of all

Earthorbitingspacecraft;4) the developmentoftechniques to distribute time through a 1553 bus; and

5) the development of low cost GPS attitude self survey

techniques using an antenna fixture. The lessons learned

from the Lewis mission are being applied to future flight

experiments, including future flights of the Tensorreceiver on the TRW ROCSAT and the GSFC EO-I

spacecraft.

Seastar

The Seastar spacecraft, developed by Orbital, includesthe Goddard SeawiFS instrument. It was launched on

August 1, 1997 and is currently in a 705 km sun-synchronous orbit. A 12 channel, 2 antenna Motorola

Viceroy receiver was flown on this mission to support

mission orbit determination and vehicle timing. The use

of GPS on this mission was not actually a flightexperiment but a mission-critical capability. Prior to the

flight, Motorola and Orbital recognized the need to fullyunderstand and characterize how the receiver would

perform in space. They requested the use of the GSFC

GPS facility[10] to independently validate the Viceroyreceiver performance. The GSFC GPS facility employs a

40 channel Nortel GPS simulator. The Viceroy was put

through a battery of tests, which proved to be critical in

improving the receiver's performance. As a result of this

testing, software modifications were made prior to flight

which have permitted the Seastar spacecraft to provide

outstanding Earth science measurements. Thus, preflighttesting of GPS in a simulator environment is crucial

when the receiver is to be used as a critical navigationdevice.

EarthWatch-Earlybird

The Earlybird-I satellite is being built by Orbital/CTA

for EarthWatch, Inc. This spacecraft represents

EarthWatch's first entry into the commercial Earth

imaging satellite business. This mission will be placed

in a 470 km, sun synchronous orbit. Of particular

interest to the GPS community is that this spacecraft will

flying two different GPS receivers--the Motorola

Viceroy and the Trimble TANS Vector C/A code

receivers. The TANS Vector, using four antennas, willprovide attitude, orbit and time data to the vehicle and

the Viceroy will provide orbit and time information to

the spacecraft. This mission has a very aggressive 3-5 m

(l-sigma) orbit accuracy requirement. A team consistingof members from Orbital, University of Colorado, and

Van Martin Systems [I I] have developed a low cost,

high accuracy Orbit Determination System for this

mission. This system employs a MicroCosm precision

orbit software system, derived from NASA Goddard'sGEODYN II, and the use of the International GPS

Service tbr Geodynamics (IGS) [12] ground system.Using this system, Davis, et. al. achieved 3-7 m orbit

accuracies using single frequency carrier phase data from

a Trimble receiver in the 800 km RADCAL [13] orbit.

The GPS team also hopes to get high accuracy attitude

data from this mission since the spacecraft includes highaccuracy attitude sensors on-board that can be used to

calibrate the GPS attitude data. The Earlybird-I

spacecraft is tentatively scheduled for launch on

December 14, 1997. The GPS collaborative experimentson this mission are sponsored by EarthWatch,

Orbital/CTA, GSFC, Trimble, Motorola, University ofColorado, Van Martin Systems and Microcosm.

Clark-GPS

In addition to the GADFLY experiment on the SSTI-

Lewis spacecraft, a GPS experiment will also be

included on the SSTI-Clark spacecraft [14], which is

currently slated to be launched in June, 1998 (see figure

9). The objectives of the Clark-GPS experiment is to

study real-time GPS orbit and attitude determination.

The experiment's physical hardware includes four GPS

antennas and pre-amplifiers, cross-strapped to twoTrimble TANS Vector GPS receivers.

Figure 9: SSTI-Clark

There are six cases of interest that will be studied with

the Clark-GPS orbit determination experiment. The GPS

Operational case will study the use of an onboard

estimator and periodic updates to improve the orbit

accuracy from the base GPS Standard Positioning

Service (SPS) accuracy of 150 meters, lo, to better than

25 meters, lo. The GPS Dropouts case will look at the

GPS data dropouts that occur in various portions of the

spacecraft orbit, particularly at high altitudes wherevisibility of GPS satellites is at its worst. Degradation of

the orbit estimator propagated orbit position during these

dropouts will also be characterized. The third case that

will be studied is the ability of the orbit estimator to

reject GPS Outliers through the use ofa chi-square test.

The rburth case to be studied for the Clark-GPS orbit

determination experiment is the performance of the GPS

receiver and the orbit propagator during and after orbit-

adjust maneuvers. The case of an inoperative GPSreceiver is also considered, with the ability on the

spacecraft to switch to the redundant GPS receiver and

reinitialize the orbit estimator by ground command.

Finally, the case of autonomous initialization of the orbitestimator after safehold will also be studied.

The Clark-GPS attitude determination experiment will

study the generation of GPS attitude for a spacecraft in

earth-pointing mode. As a long-term attitude

determination experiment, the static and dynamic error

sources of GPS attitude can be characterized by

comparing them to more accurate attitude information

from SSTI-Clark's star tracker and other spacecraft

sensors. The impact of vehicle multipath on GPS attitude

determination accuracy can also be investigated.

The Clark-GPS experiment on SSTI-Clark has been put

together by a team consisting of CTA, Welch

Engineering, and NASA Goddard Space Flight Center.

AMSAT-GPS

AMSAT Phase 3D is the latest in a long series ofsatellites built by the Radio Amateur Satellite

Corporation (AMSAT). It is typical of AMSAT satellites

in that it is being built almost entirely by a world-wide

volunteer staff of amateur radio operators and satellite

enthusiasts from AMSAT, the NASA Goddard Space

Flight Center, and others throughout the world. The

Phase 3D satellite is scheduled for launch as a secondary

payload on an Ariane 5 flight in 1998.

The AMSAT-GPS experiment will be the first to study

the use of GPS signals above the GPS constellation.AMSAT Phase 3D will be in a 4,000 x 47,000 kilometer

Molniya orbit, its apogee well over the 20,000 kilometer

altitude of the GPS satellites. In addition to providingfurther long-term, real-time GPS attitude and orbit

determination experience, the AMSAT-GPS experiment

will be able to map the GPS constellation signal patterns

available above the constellation, as well as give an

understanding of the robustness and limitations of

making use of GPS in this region.

The hardware for the AMSAT-GPS experiment willconsist of two Trimble TANS Vector GPS receivers,

along with two sets of four GPS antennas. Four patch

antennas will be located on the perigee side of the

spacecraft, while four high-gain antennas will be placed

on the apogee side.

SPACEHAB Universal Communications System(SHUCS)

The SPACEHAB Universal Communications System

(SHUCS) uses a flat panel L-band phased array antenna

mounted to a two-axis pointing system to communicate

through an lnmarsat satellite. SHUCS is mounted on topof SPACEHAB. SPACEHAB is a reusable commercial

habitation module that is flown in the Orbiter cargo bay.

This system, similar to the TAPS system described

previously, allows customers to receive 64 kbps of data

directly and send data commands in real-time; bypassingthe NASA communications network. The antenna

pointing system is controlled by a Trimble TANS Vector

navigation and attitude receiver provided by NASA

Goddard. The SHUCS experiment is planned to fly on

STS-91 in May 1998. This represents the first time aspaceborne pointing system will use GPS to steer an

antenna to a target. During the flight, GPS receiver and

SHUCS operational data will be provided to theSPACEHAB and Goddard experimenters in real-time

through the lnmarsat/SHUCS link-up. An educational

outreach program has been built into this experiment to

allow the Goddard obtained flight data to be available to

students in real time through a World Wide Web site.

SAC-A

The SAC-A

spacecraft is a small

free-flying satellite

designed by CONAEin Argentina. SAC-A

will be deployed from

a hitchhiker bridge

mounted in the SpaceShuttle Endeavour in

July 1998. The SAC-

A GPS experiment is

expected to performthe first ever attitude

and navigation

sensing on a spinning

spacecraft. As shown

Figure 10: SAC-A

in figure 10, four patch antennas with a very short

baseline (<0.5 meter) are mounted on the top of the

SAC-A spacecraft. A Trimble TANS Vector receiverwill be employed on SAC-A to determine the vehicle

attitude, orbit and time. This is a collaborative

partnership between CONAE in Argentina and theNASA Goddard Space Flight Center.

EO-I Enhanced Formation Flying Experiment

The primary objective of the enhanced formation flying

experiment on the New Millenium Program (NMP) EarthOrbiter-I (EO-I) mission is to demonstrate onboard

autonomous relative navigation and formation flying

control between the EO-1 and Landsat-7 spacecraft. An

automated mission design and automated maneuver

planning tool, AUTOCON, which was developed by AI

Solutions under direction by the Goddard GN&C team,

has been used for operational mission design.

AUTOCON is being modified to operate onboard thespacecrafl to support autonomous constellation control

and formation flying. This will be accomplished by

having the flight control system plan a maneuver that

places EO-1 within 1 minute of separation from Landsat-

7 and then maintain that separation to a tight tolerance of6 seconds for an extended period of time.

The algorithms and software tools for this demonstration

will be developed using a modular approach so that they

can easily be used onboard future Earth orbiting

missions. These algorithms [15] will be implemented

using fuzzy logic engines for constraint checking andcontrol of the formation flying algorithms.

The key benefits of this enhanced formation flying

technology are to eliminate routine ground maneuverplanning and commanding requirements, reduce costs,

enhanced science capabilities, and to advance the

technology for complete lights-out application for the

New Millennium Program. The system will provide a

real-time low-cost formation flying control with the

flexibility to meet a broad range of mission requirementsincluding ground track, inclination, and altitude control

as individual or multiple spacecraft requirements.

Space Shuttle Flights

Since 1993, engineers at the NASA Johnson Space

Center have sponsored several different experiments tovalidate the concepts required to fly GPS as an in-line

avionics component on the US Space Shuttles. The

current phase of Shuttle/GPS operations is the use of a

Precise Positioning Service (PPS) receiver as a single

string navigation device. A Rockwell Collins MAGR/S

[1][16] receiver was chosen for this phase of the

development effort. During the flights, the navigation

data is downlinked to the ground and is available in real-

time on the astronaut crew displays. During landing, the

GPS system performance will be compared to the results

obtained by TACAN, the primary Shuttle navigation

device. The single string GPS system is currently

incorporated on 3 Shuttles (Discovery, Atlantis and

Endeavour). It was first demonstrated on the STS-79

mission which was launched on September 16, 1996.

The final phase of the Shuttle GPS validation is to

incorporate a three string GPS on all four Orbiters. Since

the Shuttle avionics systems requirement is to be twofault tolerant, a minimum of three GPS receivers are

required per Space Shuttle. Once this final phase hasbeen completed and the system validated, TACANservice will be removed for future Shuttle Missions and

the GPS antennas will be installed in place of the

TACAN antennas. Currently, TACAN navigation isprovided for the Shuttle within 300 miles of the landing

site. The first flight of the GPS-only navigation solution

will be on STS-92, which is tentatively scheduled for

January 1999.

Using GPS for Space Shuttle Navigation will allow the

Shuttle to eliminate its need for TACAN navigation. In

addition, GPS is expected to improve on-orbit navigation

performance and save approximately $3-4 Million per

year by eliminating the need for Microwave Landing

System ground stations.

CONCLUSIONS

Significant systems cost, power, and weight savings, as

well as enhanced vehicle autonomy are expected from

exploiting GPS technologies in future space vehicles.

Several spacebome GPS receivers have been developedand several others are being developed to ensure that

future spacecraft and constellations of spacecraft canoperate with minimal ground intervention. A combined

industry, university and government team of partners

have defined the vision and direction for spacebome

GPS and are implementing this vision in the most costeffective manner.

REFERENCES

[1] Murray, S.V., "Global Positioning System for the

Space Shuttle", Presented at NASA GSFC, February,1996.

[2] Fuller, R.A., Kemper, B., and Rodden, J.J.,

"Spacecraft Guidance and Control with GPS TENSOR",

19 'h Annual AAS Guidance and Control Conference,

Breckenridge, CO, February 7-11, 1996.

[3] Sacchetti, A., "GPS for Orbit and Attitude

Determination: Hardware Design and Qualification Plan

for a Spaceborne Receiver," ION-GPS-94, Salt Lake

City, UT, September 20-23, 1994.

[4] Freesland,D., et. al., "GPS Based Attitude

Determination, The REX-II Experience", AIAA/USU

Small Satellite Conference, Logan, UT, September 1996.

[5] Lightsey, E.G., et al., "Flight Results of GPS Based

Attitude Control on the REX-II Spacecraft ", ION-GPS-

96, Kansas City, MO.

[6] Theroux, Y., "Open Architecture Design for GPSApplications," ION-GPS-95, Palm Springs, CA,

September 12-15, 1995.

[7] Clark, T., Varney, D. and Bauer, F., "A Low Cost

Spacecraft Receiver for Orbit and Attitude

Determination," ION-GPS-94, Salt Lake City, UT,

September 20-23, 1994.

[8] Bauer, F. H., et. al., "The GPS Attitude

Determination Flyer (GADFLY): A Space-Qualified

GPS Attitude Receiver on the SSTI-Lewis Spacecraft",

ION-GPS-95, Palm Springs, CA.

[9] Hart, R.C., et. ai. "Global Positioning System (GPS)

Enhanced Orbit Determination Experiment (GEODE) on

the Small Satellite Technology Initiative (SSTI) Lewis

Spacecraft", ION-GPS-96, Kansas City, MO.

[10] O'Donnell, J. R., Jr., et. al., "Testing of GPS-Based

Attitude Control Systems", ION-GPS-96, Kansas City,MO.

[11] Davis, G.W., et. al., "A Low Cost, High Accuracy

Automated GPS-Based Orbit Determination System for

Low Earth Satellites", ION-GPS-97, Kansas City, MO.

[12] Beutler, G., and Neilan, R., International GPS

Service for Geodynamics Resource Information, January1996.

[13] Lightsey, E.G., et al., "Application of GPS attitude

Determination to a Gravity Gradient Stabilized

Spacecraft," ION-GPS-93, Salt Lake City, UT,

September 1993.

[14] Freesland, D., et. aL, "SSTI-Clark Attitude

Determination and Control Subsystem Status Review",

CTA Design Review, January 1996.

[15] Bauer, F.H., Bristow, J., Folta, D., Hartman, K.,

How, J. and Quinn, D., "Satellite Formation Flying using

an Innovative Autonomous control System (Autocon)

Environment", AIAA GN&C Conference, New Orleans,

LA, August 1997.

[16] Kirpes, R.R., and Burkle, M.W., "Ballistic

Propagation of the MAGR-Shuttle Navigation Solution",

ION Proceedings of the 52 ndAnnual Mtg, Cambridge,MA, June 1996.